Update README.md

README.md

@@ -72,5 +72,30 @@ model-index:
       value: 45.928
       verified: true
 ---
+Trained on Samsum train split. 
 
-Details coming soon
+Parameters for training:
+no_decay = ["bias", "LayerNorm.weight", "layer_norm.weight"]
+optimizer_grouped_parameters = [
+    {
+        "params": [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)],
+        "weight_decay": 0.0,
+    },
+    {
+        "params": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)],
+        "weight_decay": 0.0,
+    },
+]
+
+lr = 0.00005
+optimizer = torch.optim.RAdam(optimizer_grouped_parameters, lr=lr)
+
+lr_scheduler = get_scheduler(
+        name="linear",
+        optimizer=optimizer,
+        num_warmup_steps=0,
+        num_training_steps=50005)
+
+This was only for 10K steps
+
+More details coming soon